首页|基于树增强朴素贝叶斯的科技计划项目风险评价体系研究

基于树增强朴素贝叶斯的科技计划项目风险评价体系研究

扫码查看
为保障科技计划项目的顺利实施并实现创新目标,以浙江省宁波、丽水、温州三市为例,构建科技计划项目风险评价体系,采用树增强朴素贝叶斯方法(tree augmented naive Bayes,TAN)训练树增强朴素贝叶斯-贝叶斯网络模型(tree augmented naive Bayes-Bayesian networks,TAN-BN),对宁波、丽水、温州三市的科技计划项目进行风险评估分析,旨在辅助政府与管理者深入理解并有效应对科技项目执行过程中的风险,以保障科技项目按计划完成并达到预定目标。研究结果显示:风险评估模型分类性能优越,有效揭示了多维度风险因素对项目风险的影响。同时,指出可建立动态风险评估和监控机制,更有效地管理和降低科技计划项目的风险,促进项目的顺利实施和目标的实现。
Risk Assessment System of Science and Technology Plan Project Based on Tree Augmented Naive Bayesian
In order to ensure the smooth implementation of science and technology plan projects and achieve innovation goals,taking Ningbo,Lishui and Wenzhou in Zhejiang Province as examples,the risk assessment system of science and technology plan projects is constructed.The tree augmented naive Bayes(TAN)method is used to train the tree augmented naive Bayes-Bayesian networks(TAN-BN)model,and the risk assessment analysis of science and technology plan projects in Ningbo,Lishui and Wenzhou is carried out.The aim is to assist the government and the administrators to understand fully and deal effectively with the risks in the implementation of science and technology projects,so as to ensure that the science and technology projects are completed as planned and achieve the intended goals.The results show that the risk assessment model has superior classification performance and effectively reveals the impact of multi-dimensional risk factors on project risk.At the same time,it is pointed out that a dynamic risk assessment and monitoring mechanism can be established to more effectively manage and reduce the risks of science and technology plan projects and promote the smooth implementation of projects and the realization of goals.

science and technology plan projectrisk assessment systemtree augmented naive Bayestree augmented naive Bayes-Bayesian networks model

兰佳豪、杜运潮、周红侠

展开 >

宁波工程学院经济与管理学院,浙江 宁波 315211

宁波工程学院 国际交流学院,浙江 宁波 315211

宁波大学海运学院,浙江 宁波 315211

科技计划项目 风险评价体系 树增强朴素贝叶斯 树增强朴素贝叶斯-贝叶斯网络模型

2024

宁波工程学院学报
宁波工程学院

宁波工程学院学报

影响因子:0.39
ISSN:1008-7109
年,卷(期):2024.36(4)